Exploratory Data Analysis For Hyrox Dataset

#install.packages("countrycode")

Basic Data Formatting

• Adding New Column named “Country” with country names expanded.

# Read the CSV file
df <- read.csv("hyrox_results_updated.csv")
# Create a new column with full country names using countrycode package
# Treat empty strings in nationality as NA
df$nationality[df$nationality == ""] <- NA
df$country <- countrycode(df$nationality, origin = "iso3c", destination = "country.name")

## Warning: Some values were not matched unambiguously: ALG, ARU, BRU, BUL, CAY, CGO, CRC, CRO, DEN, ENG, ESA, GER, GRE, GUA, INA, IRI, KOS, KSA, KUW, LAT, LIB, MAD, MAS, MON, MRI, MTN, NED, NIG, PHI, PLE, POR, PUR, RSA, SCO, SIN, SKN, SLO, SRI, SUD, SUI, TPE, TRI, UAE, URU, VIE, ZIM

# Identify unmatched values
unmatched <- unique(df$nationality[is.na(df$country)])
# Create a custom mapping for unmatched values
custom_codes <- c(
  ALG = "Algeria", ARU = "Aruba", BRU = "Brunei", BUL = "Bulgaria", CAY = "Cayman Islands",
  CGO = "Congo", CRC = "Costa Rica", CRO = "Croatia", DEN = "Denmark", ENG = "England",
  ESA = "El Salvador", GER = "Germany", GRE = "Greece", GUA = "Guatemala", INA = "Indonesia",
  IRI = "Iran", KOS = "Kosovo", KSA = "Saudi Arabia", KUW = "Kuwait", LAT = "Latvia",
  LIB = "Lebanon", MAD = "Madagascar", MAS = "Malaysia", MON = "Monaco", MRI = "Mauritius",
  MTN = "Mauritania", NED = "Netherlands", NIG = "Nigeria", PHI = "Philippines", PLE = "Palestine",
  POR = "Portugal", PUR = "Puerto Rico", RSA = "South Africa", SCO = "Scotland", SIN = "Singapore",
  SKN = "Saint Kitts and Nevis", SLO = "Slovenia", SRI = "Sri Lanka", SUD = "Sudan",
  SUI = "Switzerland", TPE = "Taiwan", TRI = "Trinidad and Tobago", UAE = "United Arab Emirates",
  URU = "Uruguay", VIE = "Vietnam", ZIM = "Zimbabwe"
)
# Fill in unmatched countries using custom mapping
missing_idx <- which(is.na(df$country) & df$nationality %in% names(custom_codes))
df$country[missing_idx] <- custom_codes[df$nationality[missing_idx]]

# Adding country column aftrer Nationality column
df <- df %>% relocate(country, .after = nationality)

• Renaming each columns so that it’s readable.

# Renaming column names for readability
colnames(df) <- c("event_id",   "event_name",   "season",   "year", "event_venue",  "gender",   "nationality",  "country",  "age_group",    "division", "total_time",   "work_time",    "total_transition_time",    "run_time", "run_1",    "ski_erg",  "transition_time_1",    "run_2",    "sled_push",    "transition_time_2",    "run_3",    "sled_pull",    "transition_time_3",    "run_4",    "burpee_broad_jumps",   "transition_time_4",    "run_5",    "rowing",   "transition_time_5",    "run_6",    "farmers_carry",    "transition_time_6",    "run_7",    "sandbag_lunges",   "transition_time_7",    "run_8",    "wall_balls",   "transition_time_8")

• Converting each time columns from HH:MM:SS to Seconds

# Convert Time Columns to Seconds
hyrox_df <- df
time_cols <- grep("time|run_|ski|sled|burpee|rowing|work_|roxzone_|carry|lunges|balls", names(df), value = TRUE)

# Strip whitespace and convert to seconds using hms()
for (col in time_cols) {
  hyrox_df[[col]] <- as.numeric(hms(as.character(trimws(df[[col]]))))
}

• Check whether the participants completed the competition. Checking each time fileds. If any of the time fields are 0, means the participant didn’t completed the competition.

# Identify all time-related columns
time_cols <- grep("total_|run_|ski|sled|burpee|rowing|work_|roxzone_|carry|lunges|balls", names(hyrox_df), value = TRUE)

# Flag those who completed all segments (no zero in time columns)
hyrox_df$completed <- ifelse(apply(hyrox_df[time_cols], 1, function(row) all(row > 0)), "Yes", "No")

# View summary
table(hyrox_df$completed)

## 
##    No   Yes 
##   418 91956

• Winner among the 3 seasons.

completed_df <- hyrox_df %>% filter(completed == "Yes")
winner_df <- completed_df %>% filter(total_time == min(total_time,na.rm = TRUE))
winner_df

##      event_id     event_name season year event_venue gender nationality country
## 1 JGDMS4JI642 S6 2023 Dallas     S6 2023      Dallas   male        <NA>    <NA>
##   age_group division total_time work_time total_transition_time run_time run_1
## 1       U40    relay       2862      1170                   178     1514   177
##   ski_erg transition_time_1 run_2 sled_push transition_time_2 run_3 sled_pull
## 1     216                20   191       112                31   184       140
##   transition_time_3 run_4 burpee_broad_jumps transition_time_4 run_5 rowing
## 1                34   184                 78                33   192    223
##   transition_time_5 run_6 farmers_carry transition_time_6 run_7 sandbag_lunges
## 1                17   194            66                22   194            128
##   transition_time_7 run_8 wall_balls transition_time_8 completed
## 1                21   198        207                 0       Yes

• Countries with more than 500 participants among 3 seasons.

countries_to_view <- hyrox_df %>% count(nationality) %>% filter(n > 500) %>% pull(nationality)
countries_to_view <- countries_to_view[-1]
filtered_df <- hyrox_df %>% filter(nationality %in% countries_to_view) %>% filter(division == "open")
countries_to_view

##  [1] "AUT" "ESP" "FRA" "GBR" "GER" "HKG" "IRL" "ITA" "NED" "SIN" "SUI" "SWE"
## [13] "USA" NA

• Create new data frame filtered_df with participants from top countries with division == “open”.

countries_to_view <- hyrox_df %>% count(country) %>% filter(n > 500)%>% filter(!is.na(country)) %>% pull(country)
filtered_df <- hyrox_df %>% filter(country %in% countries_to_view) %>% filter(division == "open")

glimpse(filtered_df)

## Rows: 47,700
## Columns: 39
## $ event_id              <chr> "JGDMS4JI5C9", "JGDMS4JI5C9", "JGDMS4JI5C9", "JG…
## $ event_name            <chr> "S6 2023 München", "S6 2023 München", "S6 2023 M…
## $ season                <chr> "S6", "S6", "S6", "S6", "S6", "S6", "S6", "S6", …
## $ year                  <int> 2023, 2023, 2023, 2023, 2023, 2023, 2023, 2023, …
## $ event_venue           <chr> "München", "München", "München", "München", "Mün…
## $ gender                <chr> "male", "male", "male", "male", "male", "male", …
## $ nationality           <chr> "GER", "AUT", "GER", "GER", "FRA", "GER", "GER",…
## $ country               <chr> "Germany", "Austria", "Germany", "Germany", "Fra…
## $ age_group             <chr> "40-44", "25-29", "30-34", "30-34", "30-34", "25…
## $ division              <chr> "open", "open", "open", "open", "open", "open", …
## $ total_time            <dbl> 3547, 3567, 3793, 3840, 3949, 3974, 4013, 4017, …
## $ work_time             <dbl> 1531, 1506, 1649, 1741, 1724, 1629, 1822, 1655, …
## $ total_transition_time <dbl> 206, 177, 194, 202, 202, 301, 222, 284, 197, 225…
## $ run_time              <dbl> 1810, 1884, 1950, 1897, 2023, 2044, 1969, 2078, …
## $ run_1                 <dbl> 192, 195, 208, 202, 211, 194, 212, 211, 237, 245…
## $ ski_erg               <dbl> 246, 239, 245, 253, 245, 231, 269, 250, 270, 239…
## $ transition_time_1     <dbl> 2, 2, 1, 2, 2, 2, 2, 2, 2, 3, 1, 2, 2, 2, 4, 3, …
## $ run_2                 <dbl> 216, 218, 232, 227, 240, 238, 238, 249, 264, 254…
## $ sled_push             <dbl> 127, 133, 142, 162, 119, 153, 201, 141, 212, 172…
## $ transition_time_2     <dbl> 27, 23, 26, 25, 27, 40, 42, 41, 27, 30, 39, 38, …
## $ run_3                 <dbl> 236, 234, 243, 248, 256, 264, 251, 259, 255, 265…
## $ sled_pull             <dbl> 200, 195, 189, 212, 243, 216, 211, 211, 241, 198…
## $ transition_time_3     <dbl> 35, 31, 32, 36, 31, 42, 34, 47, 33, 36, 36, 47, …
## $ run_4                 <dbl> 231, 240, 244, 238, 257, 265, 249, 264, 256, 275…
## $ burpee_broad_jumps    <dbl> 198, 159, 227, 150, 157, 245, 192, 201, 179, 193…
## $ transition_time_4     <dbl> 39, 36, 36, 39, 40, 53, 39, 60, 33, 44, 50, 61, …
## $ run_5                 <dbl> 228, 242, 246, 244, 263, 268, 253, 263, 256, 275…
## $ rowing                <dbl> 253, 243, 250, 275, 256, 259, 278, 262, 266, 243…
## $ transition_time_5     <dbl> 28, 19, 23, 23, 21, 34, 21, 31, 25, 28, 33, 36, …
## $ run_6                 <dbl> 230, 234, 247, 237, 264, 271, 251, 266, 261, 288…
## $ farmers_carry         <dbl> 75, 81, 102, 120, 124, 96, 100, 114, 108, 89, 12…
## $ transition_time_6     <dbl> 40, 36, 43, 45, 41, 84, 45, 53, 41, 47, 79, 40, …
## $ run_7                 <dbl> 227, 239, 251, 237, 254, 261, 254, 259, 263, 285…
## $ sandbag_lunges        <dbl> 185, 204, 217, 234, 228, 184, 266, 197, 242, 237…
## $ transition_time_7     <dbl> 35, 30, 33, 32, 40, 46, 39, 50, 36, 37, 42, 39, …
## $ run_8                 <dbl> 250, 282, 279, 264, 278, 283, 261, 307, 277, 316…
## $ wall_balls            <dbl> 247, 252, 277, 335, 352, 245, 305, 279, 289, 280…
## $ transition_time_8     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ completed             <chr> "Yes", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes",…

• Count of participants per country and season

# Count participants per country per season
country_counts_by_season <- filtered_df %>%
  filter(!is.na(country)) %>%
  count(season, country, sort = TRUE) %>% 
  arrange(country,season)

# View result
print(country_counts_by_season)

##    season             country    n
## 1      S4           Australia    1
## 2      S5           Australia   22
## 3      S6           Australia 1141
## 4      S4             Austria  194
## 5      S5             Austria  322
## 6      S6             Austria  116
## 7      S4              France   42
## 8      S5              France  251
## 9      S6              France 1139
## 10     S4             Germany 2813
## 11     S5             Germany 5058
## 12     S6             Germany 2433
## 13     S5 Hong Kong SAR China  460
## 14     S6 Hong Kong SAR China  203
## 15     S4             Ireland    8
## 16     S5             Ireland  264
## 17     S6             Ireland  788
## 18     S4               Italy   21
## 19     S5               Italy  149
## 20     S6               Italy  806
## 21     S4         Netherlands  455
## 22     S5         Netherlands 1902
## 23     S6         Netherlands 1372
## 24     S5           Singapore   22
## 25     S6           Singapore  567
## 26     S4               Spain  278
## 27     S5               Spain 1654
## 28     S6               Spain 1290
## 29     S4              Sweden    1
## 30     S5              Sweden  166
## 31     S6              Sweden  361
## 32     S4         Switzerland   39
## 33     S5         Switzerland  267
## 34     S6         Switzerland  137
## 35     S4      United Kingdom 2430
## 36     S5      United Kingdom 7589
## 37     S6      United Kingdom 6006
## 38     S4       United States 1212
## 39     S5       United States 3293
## 40     S6       United States 2428

Data Analysis Plots

Plot 1.

Below plot provides a comparative view of the number of HYROX race participants across different countries over three seasons (S4, S5, and S6).

# Create ggplot with custom tooltip
p <- ggplot(country_counts_by_season, aes(x = country, y = n, fill = season,
                                          text = paste("Country:", country,
                                                       "<br>Season:", season,
                                                       "<br>Participants:", n))) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(title = "Participant Count by Country and Season",
       x = "Country", y = "Count") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Convert to interactive plot
ggplotly(p, tooltip = "text")

The United States, Germany and United Kingdom have the highest participation across all seasons, with highest participation in Season 5 (S5). A sharp increase in participation from Season 4 to Season 5 is visible in most countries, indicating growing popularity. Season 6 (S6) shows a slight dip in some countries, possibly due to missing data.
Season 4 (2021 - 2022). Season 5 (2022 - 2023). Season 6 (2023).

Plot 2.

Below plot examines the distribution of male and female participants across three HYROX seasons (S4, S5, S6).

# Count participants per season and gender
season_gender_counts <- hyrox_df %>%
  filter(!is.na(season), !is.na(gender)) %>%
  count(season, gender)

# Create grouped bar chart
p4 <- ggplot(season_gender_counts, aes(x = season, y = n, fill = gender,
                                       text = paste("Season:", season,
                                                    "<br>Gender:", gender,
                                                    "<br>Participants:", n))) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(title = "Participants by Season and Gender", x = "Season", y = "Count") +
  theme_minimal()

# Make it interactive
ggplotly(p4, tooltip = "text")

Male participation is consistently higher than female across all seasons. - Both genders saw a significant rise in Season 5, indicating a surge in event popularity. - Season 6 experienced a slight dip compared to Season 5.

Creating new Data Frame

Creating new data frame s5_filtered_df with records for season 5 and division open. Filtering data for countries with more than 100 participants.

s5_countries_to_view <- hyrox_df %>% filter(season == "S5") %>% count(country) %>% filter(n > 100)%>% filter(!is.na(country)) %>% pull(country)
s5_filtered_df <- hyrox_df %>% filter(season == "S5", country %in% s5_countries_to_view) %>% filter(division == "open")

glimpse(s5_filtered_df)

## Rows: 21,603
## Columns: 39
## $ event_id              <chr> "2EFMS4JI2BF", "2EFMS4JI2BF", "2EFMS4JI2BF", "2E…
## $ event_name            <chr> "S5 2022 Basel", "S5 2022 Basel", "S5 2022 Basel…
## $ season                <chr> "S5", "S5", "S5", "S5", "S5", "S5", "S5", "S5", …
## $ year                  <int> 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022, …
## $ event_venue           <chr> "Basel", "Basel", "Basel", "Basel", "Basel", "Ba…
## $ gender                <chr> "male", "male", "male", "male", "male", "male", …
## $ nationality           <chr> "GBR", "GBR", "GER", "GER", "GER", "GBR", "GER",…
## $ country               <chr> "United Kingdom", "United Kingdom", "Germany", "…
## $ age_group             <chr> "30-34", "30-34", "30-34", "25-29", "35-39", "16…
## $ division              <chr> "open", "open", "open", "open", "open", "open", …
## $ total_time            <dbl> 3509, 3509, 3552, 3693, 3727, 3768, 3779, 3928, …
## $ work_time             <dbl> 1465, 1465, 1475, 1530, 1537, 1502, 1713, 1634, …
## $ total_transition_time <dbl> 244, 244, 264, 282, 294, 298, 264, 272, 319, 312…
## $ run_time              <dbl> 1800, 1800, 1813, 1881, 1896, 1968, 1802, 2022, …
## $ run_1                 <dbl> 219, 219, 230, 214, 224, 233, 203, 215, 219, 230…
## $ ski_erg               <dbl> 229, 229, 241, 237, 248, 228, 245, 245, 234, 256…
## $ transition_time_1     <dbl> 29, 29, 31, 31, 31, 31, 28, 29, 31, 32, 37, 31, …
## $ run_2                 <dbl> 218, 218, 219, 226, 231, 239, 211, 226, 236, 233…
## $ sled_push             <dbl> 152, 152, 142, 147, 113, 103, 143, 124, 126, 160…
## $ transition_time_2     <dbl> 28, 28, 30, 33, 40, 34, 31, 33, 43, 40, 41, 40, …
## $ run_3                 <dbl> 220, 220, 225, 227, 240, 251, 223, 254, 259, 243…
## $ sled_pull             <dbl> 168, 168, 185, 210, 202, 156, 231, 204, 210, 208…
## $ transition_time_3     <dbl> 38, 38, 39, 41, 43, 42, 40, 39, 50, 46, 47, 47, …
## $ run_4                 <dbl> 222, 222, 221, 235, 238, 246, 234, 255, 256, 242…
## $ burpee_broad_jumps    <dbl> 132, 132, 145, 122, 185, 193, 178, 196, 167, 191…
## $ transition_time_4     <dbl> 30, 30, 41, 36, 37, 42, 36, 34, 43, 40, 43, 36, …
## $ run_5                 <dbl> 221, 221, 227, 231, 234, 276, 231, 258, 262, 244…
## $ rowing                <dbl> 236, 236, 250, 253, 291, 248, 256, 259, 260, 258…
## $ transition_time_5     <dbl> 27, 27, 29, 32, 34, 29, 29, 30, 33, 36, 37, 33, …
## $ run_6                 <dbl> 225, 225, 221, 238, 233, 247, 227, 259, 256, 254…
## $ farmers_carry         <dbl> 100, 100, 91, 101, 85, 90, 106, 126, 101, 87, 95…
## $ transition_time_6     <dbl> 48, 48, 49, 57, 54, 59, 51, 58, 62, 58, 64, 66, …
## $ run_7                 <dbl> 229, 229, 225, 246, 234, 226, 227, 267, 259, 247…
## $ sandbag_lunges        <dbl> 182, 182, 170, 191, 184, 221, 202, 199, 221, 229…
## $ transition_time_7     <dbl> 44, 44, 45, 52, 55, 61, 49, 49, 57, 60, 53, 56, …
## $ run_8                 <dbl> 246, 246, 245, 264, 262, 250, 246, 288, 276, 285…
## $ wall_balls            <dbl> 266, 266, 251, 269, 229, 263, 352, 281, 267, 288…
## $ transition_time_8     <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ completed             <chr> "Yes", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes",…

Plot 3.

This bar chart highlights the number of participants from each country during Season 5 of the HYROX competition.

# Count participants by country
country_counts <- s5_filtered_df %>%
  filter(!is.na(country)) %>%
  count(country)

# Plot
p1 <- ggplot(country_counts, aes(x = reorder(country, -n), y = n,
                                 text = paste("Country:", country,
                                              "<br>Participants:", n))) +
  geom_bar(stat = "identity", fill = "steelblue") +
  labs(title = "Season 5: Participants by Country", x = "Country", y = "Count") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

ggplotly(p1, tooltip = "text")

The United Kingdom leads with the highest number of participants in Season 5, followed closely by Germany and the United States.

Plot 4.

Below plot presents a breakdown of HYROX participants count by gender for Season 5.

gender_counts <- s5_filtered_df %>%
  filter(!is.na(gender)) %>%
  count(gender)

p2 <- ggplot(gender_counts, aes(x = gender, y = n,
                                text = paste("Gender:", gender,
                                             "<br>Participants:", n))) +
  geom_bar(stat = "identity", fill = "darkorange") +
  labs(title = "Season 5: Participants by Gender", x = "Gender", y = "Count") +
  theme_minimal()

ggplotly(p2, tooltip = "text")

Male participants clearly outnumber female participants in Season 5.

Plot 5.

Below bar chart illustrates the distribution of participants across different age groups in Season 5 of the HYROX competition.

age_counts <- s5_filtered_df %>%
  filter(!is.na(age_group), age_group != "") %>%
  count(age_group)

p3 <- ggplot(age_counts, aes(x = age_group, y = n,
                             text = paste("Age Group:", age_group,
                                          "<br>Participants:", n))) +
  geom_bar(stat = "identity", fill = "mediumseagreen") +
  labs(title = "Season 5: Participants by Age Group", x = "Age Group", y = "Count") +
  theme_minimal()

ggplotly(p3, tooltip = "text")

The 30–34 and 35–39 age groups have the highest participation, indicating peak engagement among young to mid-career adults. - Participation steadily declines with increasing age, particularly beyond the 50–54 group.

Plot 6.

Below stacked bar chart visualizes the number of male and female participants from each country during Season 5 of the HYROX event.

country_gender_counts <- s5_filtered_df %>%
  filter(!is.na(country), !is.na(gender)) %>%
  count(country, gender)

p4 <- ggplot(country_gender_counts, aes(x = country, y = n, fill = gender,
                                        text = paste("Country:", country,
                                                     "<br>Gender:", gender,
                                                     "<br>Participants:", n))) +
  geom_bar(stat = "identity") +
  labs(title = "Season 5: Participants by Country and Gender (Stacked)",
       x = "Country", y = "Count") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

ggplotly(p4, tooltip = "text")

The United Kingdom, Germany and United States had the highest overall participation in Season 5. - In most countries, male participation significantly exceeds female participation.

Top 10 participants in Season 5

# Ensure total_time is numeric
s5_filtered_df <- s5_filtered_df %>%
  mutate(total_time = as.numeric(total_time))

# Get top 10 finishers who completed the race
top10_s5 <- s5_filtered_df %>%
  filter(completed == "Yes",                
         !is.na(total_time), total_time > 0) %>%
  arrange(total_time) %>%
  slice(1:10) %>%
  select(country, gender, age_group, total_time, event_name, nationality)

# View the result
top10_s5

##           country gender age_group total_time
## 1          France   male     35-39       3294
## 2         Germany   male     35-39       3340
## 3           Spain   male     35-39       3394
## 4           Spain   male     35-39       3394
## 5  United Kingdom   male     30-34       3424
## 6  United Kingdom   male     16-24       3429
## 7         Germany   male     16-24       3438
## 8         Germany   male     16-24       3438
## 9         Germany   male     30-34       3438
## 10        Germany   male     25-29       3467
##                                   event_name nationality
## 1                             S5 2023 Bilbao         FRA
## 2                              S5 2022 Essen         GER
## 3                              S5 2022 Basel         ESP
## 4                          S5 2023 Barcelona         ESP
## 5  S5 2023 Maastricht European Championships         GBR
## 6                            S5 2023 Glasgow         GBR
## 7                            S5 2023 Hamburg         GER
## 8                            S5 2023 Hamburg         GER
## 9  S5 2023 Maastricht European Championships         GER
## 10                              S5 2023 Wien         GER

Plot 7.

Below bar chart presents the top 10 male and top 10 female finishers based on their total time (in seconds) during Season 5.The age group of each athlete is also displayed for additional context.

# Ensure total_time is numeric
s5_filtered_df <- s5_filtered_df %>%
  mutate(total_time = as.numeric(total_time))

# Get Top 10 per gender (only completed)
top10_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes",            
         !is.na(total_time), total_time > 0) %>%
  group_by(gender) %>%
  arrange(total_time) %>%
  slice(1:10) %>%
  mutate(rank = row_number()) %>%
  ungroup()

p <- ggplot(top10_by_gender, aes(
  x = reorder(paste(gender, rank, paste0("(", age_group, ")")), -total_time),
  y = total_time, fill = gender,
  text = paste("Gender:", gender,
               "<br>Rank:", rank,
               "<br>Age Group:", age_group,
               "<br>Time:", total_time,
               "<br>Country:", country))) +
  geom_col() +
  coord_flip() +
  labs(title = "Top 10 Male and Female Performers (Season 5)",
       x = "Gender + Rank (Age Group)", y = "Total Time (seconds)") +
  theme_minimal()

ggplotly(p, tooltip = "text")

The top 10 male athletes have faster times than the top 10 female athletes. - Most top male finishers belong to the 16-24 and 35-39 age groups. - Female top performers are spread across 25-29 and 30-34 age groups.

Plot 8.

Below line chart visualizes the progression of total finishing times for the top 10 male and female participants in Season 5, ranked from fastest (1) to 10th place.

# Prepare top 10 by gender
top10_by_gender <- s5_filtered_df %>%
  mutate(total_time = as.numeric(total_time)) %>%
  filter(completed == "Yes",                # Include only completed participants
         !is.na(total_time), total_time > 0) %>%
  group_by(gender) %>%
  arrange(total_time) %>%
  slice(1:10) %>%
  mutate(rank = row_number()) %>%
  ungroup()

p1 <- ggplot(top10_by_gender, aes(x = rank, y = total_time, color = gender,
                                  text = paste("Gender:", gender,
                                               "<br>Rank:", rank,
                                               "<br>Time:", total_time,
                                               "<br>Country:", country))) +
  geom_line(aes(group = gender)) +
  geom_point(size = 3) +
  scale_x_continuous(breaks = 1:10) +  # This sets x-axis ticks to 1–10
  labs(title = "Top 10 Total Time by Rank (Male vs Female) - Season 5", 
       x = "Rank", y = "Total Time (seconds)") +
  theme_minimal()

ggplotly(p1, tooltip = "text")

Male participants consistently had faster finishing times compared to female participants across all ranks. - The gap between male and female times remains steady, with female athletes taking approximately 300–400 seconds longer.

Plot 9.

Below plot shows how finishing times of the fastest athlete and the average of the top 10 performers, segmented by gender, have evolved across three seasons (S4, S5, S6).

# Ensuring total_time is numeric and filter valid + completed entries
filtered_df <- filtered_df %>%
  mutate(total_time = as.numeric(total_time)) %>%
  filter(completed == "Yes",                  # Include only completed
         !is.na(total_time), total_time > 0)

# Finding fastest performer per season by gender
fastest_by_gender <- filtered_df %>%
  group_by(season, gender) %>%
  slice_min(total_time, n = 1) %>%
  mutate(metric = "Fastest") %>%
  select(season, gender, total_time, metric, country)

# Average of top 10 per season and gender
avg_top10_by_gender <- filtered_df %>%
  group_by(season, gender) %>%
  arrange(total_time) %>%
  slice(1:10) %>%
  summarise(total_time = mean(total_time),
            country = NA,  # placeholder to match columns
            .groups = "drop") %>%
  mutate(metric = "Avg Top 10")

# Combine both metrics
top_perf_all <- bind_rows(fastest_by_gender, avg_top10_by_gender)

# Plot the graph
p <- ggplot(top_perf_all, aes(x = season, y = total_time,
                              color = gender, linetype = metric,
                              group = interaction(gender, metric),
                              text = paste("Season:", season,
                                           "<br>Gender:", gender,
                                           "<br>Metric:", metric,
                                           "<br>Total Time:", round(total_time, 2),
                                           ifelse(is.na(country), "", paste("<br>Country:", country))))) +
  geom_line(size = 1.2) +
  geom_point(size = 3) +
  labs(title = "Seasonal Performance: Fastest vs Avg Top 10 (by Gender)",
       x = "Season", y = "Total Time (seconds)", color = "Gender", linetype = "Metric") +
  theme_minimal()

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

ggplotly(p, tooltip = "text")

Male participants consistently outperformed females in both fastest time and average top 10 time across all seasons. - Both male and female fastest times improved from Season 4 to Season 5, with males showing a more substantial gain. - Season 6 saw a slight performance drop for females (in both metrics), while male times remained relatively stable.

Plot 10.

Below density plot shows the distribution of total finishing times for male and female participants in Season 5.

df1 <- s5_filtered_df %>%
  filter(completed == "Yes", !is.na(gender))
p4 <- ggplot(df1, aes(x = total_time, fill = gender)) +
  geom_density(alpha = 0.5) +
  labs(title = "Total Time Density by Gender - Season 5", x = "Total Time (seconds)", y = "Density") +
  theme_minimal()

ggplotly(p4)

Both male and female distributions are right-skewed, indicating that most participants finish within a certain range, with fewer taking significantly longer. - The plot suggests that males generally completed the race in lesser time.

Plot 11.

Below bar chart presents the average total finishing time (in seconds) of participants who completed the race, grouped by country in Season 5.

# Calculate average total_time by country (only for completed participants)
avg_time_by_country <- s5_filtered_df %>%
  filter(completed == "Yes", !is.na(country), !is.na(total_time), total_time > 0) %>%
  group_by(country) %>%
  summarise(avg_total_time = mean(total_time), .groups = "drop")

# Plot bar chart of average total_time
p_avg <- ggplot(avg_time_by_country, aes(x = reorder(country, avg_total_time),
                                         y = avg_total_time,
                                         text = paste("Country:", country,
                                                      "<br>Avg Time:", round(avg_total_time, 2)))) +
  geom_col(fill = "steelblue") +
  labs(title = "Average Total Time by Country (Completed Participants) - Season 5",
       x = "Country", y = "Average Total Time (seconds)") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Make it interactive
ggplotly(p_avg, tooltip = "text")

Portugal has the lowest average finishing time, suggesting a generally faster performance by its participants. - Most European countries, such as France, Italy, Spain, Germany, and Switzerland, have similar average times, clustering around the middle of the performance spectrum.

Plot 12.

The below plot shows the average total finishing time (in seconds) for participants across different age groups in Season 5.

# Set desired age group order manually (as character strings)
age_levels <- c("16-24", "25-29", "30-34", "35-39", "40-44", "45-49", 
                "50-54", "55-59", "60-64", "65-69", "70-74", "75-79", "80-84")

avg_time_by_age <- filtered_df %>%
  filter(completed == "Yes", !is.na(age_group), !is.na(total_time), total_time > 0) %>%
  group_by(age_group) %>%
  summarise(avg_total_time = mean(total_time), .groups = "drop") %>%
  mutate(age_group = factor(age_group, levels = age_levels, ordered = TRUE))

# Plot
p <- ggplot(avg_time_by_age, aes(x = age_group, y = avg_total_time,
                                 fill = age_group,
                                 text = paste("Age Group:", age_group,
                                              "<br>Avg Time:", round(avg_total_time, 2)))) +
  geom_col(show.legend = FALSE) +  # remove redundant legend
  labs(title = "Average Total Time by Age Group (Completed Participants)",
       x = "Age Group", y = "Average Total Time (seconds)") +
  theme_minimal()

ggplotly(p, tooltip = "text")

Participants aged 16–24 to 35–39 show relatively lower average finishing times, indicating stronger performance among younger competitors. - A gradual increase in average total time is observed starting from age group 40–44 onwards. - Participants in the 65–69 and 70+ age brackets have noticeably higher average times, which is expected due to age-related endurance variation.

Plot 13.

The below plot provides a comparison of the average time (in seconds) spent on each event segment by two groups in Season 5: - All completed participants - Top 10 performers

# Time segments
all_time_cols <- grep("run_|ski|sled|burpee|row|carry|lunges|balls|transition_", 
                      names(filtered_df), value = TRUE)

segment_labels <- all_time_cols %>%
  gsub("_", " ", .) %>%
  tools::toTitleCase()

# Base dataset for completed participants
df_all <- filtered_df %>%
  filter(completed == "Yes", !is.na(total_time)) %>%
  select(total_time, all_of(all_time_cols)) %>%
  mutate(group = "All")

# Get top 10 performers
df_top <- df_all %>%
  arrange(total_time) %>%
  slice(1:10) %>%
  mutate(group = "Top 10")

# Combine both
df_combined <- bind_rows(df_all, df_top)

# Pivot to long format
df_long <- df_combined %>%
  pivot_longer(cols = all_of(all_time_cols), names_to = "segment_raw", values_to = "time_sec") %>%
  filter(!is.na(time_sec), time_sec > 0) %>%
  mutate(segment = factor(segment_raw, levels = all_time_cols, labels = segment_labels))

# Remove overall totals
df_long <- df_long %>%
  filter(!segment %in% c("Run Time", "Total Transition Time"))

# Compute averages
avg_segment_group <- df_long %>%
  group_by(group, segment) %>%
  summarise(avg_time = mean(time_sec), .groups = "drop")

# Plot grouped bar chart
p <- ggplot(avg_segment_group, aes(x = segment, y = avg_time, fill = group,
                                   text = paste("Group:", group,
                                                "<br>Segment:", segment,
                                                "<br>Avg Time:", round(avg_time, 2)))) +
  geom_col(position = "dodge") +
  labs(title = "Segment-wise Average Time: All vs Top Performers",
       x = "Segment", y = "Average Time (seconds)", fill = "Group") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))  # ✅ Fix overlap


ggplotly(p, tooltip = "text")

Top performers consistently spent less time on almost every segment, including both runs and workout stations. - Even transition times are significantly shorter for top performers, showcasing their efficiency in moving between stations.

Plot 14.

Below plot showcases the segment-wise performance times of the top performer from each country during Season 5.

# Identify time segment columns (excluding 'total_transition_time' and 'run_time')
segment_cols <- grep("run_|ski|sled|burpee|row|carry|lunges|balls|transition_", 
                     names(s5_filtered_df), value = TRUE)
segment_cols <- setdiff(segment_cols, c("total_transition_time", "run_time"))

# Top performer from each country (completed only)
top_by_country <- s5_filtered_df %>%
  filter(completed == "Yes", !is.na(country), !is.na(total_time), total_time > 0) %>%
  group_by(country) %>%
  slice_min(order_by = total_time, n = 1) %>%
  ungroup() %>%
  select(country, all_of(segment_cols))

# Pivot to long format
top_segment_long <- top_by_country %>%
  pivot_longer(cols = -country, names_to = "segment_raw", values_to = "time_sec") %>%
  filter(!is.na(time_sec), time_sec > 0)

# Clean segment names and preserve order
segment_labels <- segment_cols %>%
  gsub("_", " ", .) %>%
  tools::toTitleCase()

top_segment_long <- top_segment_long %>%
  mutate(segment = factor(segment_raw, levels = segment_cols, labels = segment_labels))

# Line chart with thin lines
p <- ggplot(top_segment_long, aes(x = segment, y = time_sec, color = country, group = country,
                                  text = paste("Country:", country,
                                               "<br>Segment:", segment,
                                               "<br>Time:", round(time_sec, 2)))) +
  geom_line(linewidth = 0.7) +
  geom_point(size = 2) +
  labs(title = "Top Performer Segment Times by Country",
       x = "Segment", y = "Time (seconds)", color = "Country") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Interactive plot
ggplotly(p, tooltip = "text")

The plot shows variation in segment times across top performers from different countries. While run segments are generally consistent, there is noticeable disparity in workout-intensive sections like Sled Push, Sandbag Lunges, and Farmers Carry. Some athletes demonstrate particularly fast transitions, while others lose time in these areas.

Plot 15.

The below plot compares the average run segment times across Season 5 for different profiles: average male, average female, overall average, and the top-performing male and female athletes.

# Run segment columns (excluding 'run_time')
run_cols <- grep("^run_", names(s5_filtered_df), value = TRUE)
run_cols <- setdiff(run_cols, "run_time")

# Gender-wise average
avg_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  summarise(across(all_of(run_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec")

# Overall average
avg_overall <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  summarise(across(all_of(run_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(everything(), names_to = "segment", values_to = "time_sec") %>%
  mutate(group = "Overall Avg")

# Top performer by gender
top_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  slice_min(total_time, n = 1) %>%
  ungroup() %>%
  select(gender, all_of(run_cols)) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec") %>%
  mutate(group = paste("Top", gender))

# Combine all
combined_run <- bind_rows(
  avg_by_gender %>% mutate(group = paste(gender, "Avg")),
  top_by_gender,
  avg_overall
) %>%
  mutate(segment = gsub("_", " ", segment),
         segment = tools::toTitleCase(segment),
         segment = factor(segment, levels = unique(segment)))

# Plot with linetype
p_run <- ggplot(combined_run, aes(x = segment, y = time_sec,
                                  color = group, linetype = group, group = group,
                                  text = paste("Group:", group,
                                               "<br>Segment:", segment,
                                               "<br>Time:", round(time_sec, 2)))) +
  geom_line(linewidth = 0.7) +
  geom_point(size = 2) +
  labs(title = "Run Segment Comparison (Season 5)",
       x = "Run Segment", y = "Time (seconds)",
       color = "Profile", linetype = "Profile") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Interactive plot
ggplotly(p_run, tooltip = "text")

The top male performers consistently recorded the fastest times across all run segments, followed by top females. Average male and female times were notably higher, with female participants having the longest run times overall. A noticeable upward trend in times is seen toward later run segments (e.g., Run 8).

Plot 16.

Below plot compares the average times taken by different participant groups—average male, average female, overall average, and top-performing male and female athletes—across all major exercise segments in Season 5.

# Define exercise segment columns
exercise_cols <- grep("ski|sled|burpee|row|carry|lunges|balls", names(s5_filtered_df), value = TRUE)

# Gender-wise average
avg_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  summarise(across(all_of(exercise_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec")

# Overall average
avg_overall <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  summarise(across(all_of(exercise_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(everything(), names_to = "segment", values_to = "time_sec") %>%
  mutate(group = "Overall Avg")

# Top performer by gender
top_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  slice_min(total_time, n = 1) %>%
  ungroup() %>%
  select(gender, all_of(exercise_cols)) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec") %>%
  mutate(group = paste("Top", gender))

# Combine all
combined_ex <- bind_rows(
  avg_by_gender %>% mutate(group = paste(gender, "Avg")),
  top_by_gender,
  avg_overall
) %>%
  mutate(segment = gsub("_", " ", segment),
         segment = tools::toTitleCase(segment),
         segment = factor(segment, levels = unique(segment)))

# Plot with linetype
p_exercise <- ggplot(combined_ex, aes(x = segment, y = time_sec,
                                      color = group, linetype = group, group = group,
                                      text = paste("Group:", group,
                                                   "<br>Segment:", segment,
                                                   "<br>Time:", round(time_sec, 2)))) +
  geom_line(linewidth = 0.7) +
  geom_point(size = 2) +
  labs(title = "Exercise Segment Comparison (Season 5)",
       x = "Exercise Segment", y = "Time (seconds)",
       color = "Profile", linetype = "Profile") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

ggplotly(p_exercise, tooltip = "text")

Top performers of both genders completed each exercise segment considerably faster than the average participants. In most segments, top male performers recorded the shortest times. However, Sled Push and Wall Balls are exceptions, where the top female performer outpaced her male counterpart. Notably, for these two segments, the top female’s times were also lower than the male average, female average, and overall average—highlighting a standout performance.

Plot 17.

Below plot compares the average time taken in each transition segment during Season 5 across different profiles: top male performers, top female performers, male average, female average, and overall average. Transitions refer to the periods between workout and run segments, and this breakdown helps assess how efficiently participants manage these interludes.

# Define transition columns and remove 'total_transition_time' and 'transition_time_8'
transition_cols <- grep("transition_", names(s5_filtered_df), value = TRUE)
transition_cols <- setdiff(transition_cols, c("total_transition_time", "transition_time_8"))

# Gender-wise average
avg_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  summarise(across(all_of(transition_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec")

# Overall average
avg_overall <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  summarise(across(all_of(transition_cols), ~ mean(.x, na.rm = TRUE))) %>%
  pivot_longer(everything(), names_to = "segment", values_to = "time_sec") %>%
  mutate(group = "Overall Avg")

# Top performer by gender
top_by_gender <- s5_filtered_df %>%
  filter(completed == "Yes") %>%
  group_by(gender) %>%
  slice_min(total_time, n = 1) %>%
  ungroup() %>%
  select(gender, all_of(transition_cols)) %>%
  pivot_longer(-gender, names_to = "segment", values_to = "time_sec") %>%
  mutate(group = paste("Top", gender))

# Combine all
combined_trans <- bind_rows(
  avg_by_gender %>% mutate(group = paste(gender, "Avg")),
  top_by_gender,
  avg_overall
) %>%
  mutate(segment = gsub("_", " ", segment),
         segment = tools::toTitleCase(segment),
         segment = factor(segment, levels = unique(segment)))

# Plot with linetype
p_transition <- ggplot(combined_trans, aes(x = segment, y = time_sec,
                                           color = group, linetype = group, group = group,
                                           text = paste("Group:", group,
                                                        "<br>Segment:", segment,
                                                        "<br>Time:", round(time_sec, 2)))) +
  geom_line(linewidth = 0.7) +
  geom_point(size = 2) +
  labs(title = "Transition Segment Comparison (Season 5)",
       x = "Transition Segment", y = "Time (seconds)",
       color = "Profile", linetype = "Profile") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Interactive plot
ggplotly(p_transition, tooltip = "text")

Top performers from both genders consistently completed transition segments faster than the average participants. Top male performers exhibited the lowest transition times overall, often outperforming both the top female performers and all averages. While female averages were slightly higher across most segments, the gap between top and average performers is notably larger, especially from Transition Time 3 onward, indicating better pacing and time management among top-tier athletes.

Plot 18.

Below boxplot visualizes the distribution of total finishing times for male and female participants across different countries. Each box represents the interquartile range (IQR), with the horizontal line indicating the median, and the whiskers extending to the non-outlier range. Outliers are shown as individual points.

# Filter dataset
df <- s5_filtered_df %>%
  filter(completed == "Yes", !is.na(country), !is.na(gender), !is.na(total_time))

# Create improved boxplot
p_gender_box <- ggplot(df, aes(x = country, y = total_time,
                               fill = gender, color = gender, group = interaction(country, gender),
                               text = paste("Country:", country,
                                            "<br>Gender:", gender,
                                            "<br>Time (sec):", round(total_time, 2)))) +
  geom_boxplot(position = position_dodge(width = 0.75),
               alpha = 0.6, outlier.alpha = 0.2) +
  labs(title = "Finishing Time by Gender Across Countries",
       x = "Country", y = "Total Time (seconds)",
       fill = "Gender", color = "Gender") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Make it interactive
ggplotly(p_gender_box, tooltip = "text")

## Warning: The following aesthetics were dropped during statistical transformation: text.
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
##   the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
##   variable into a factor?

Across the majority of countries, male participants generally outperform female participants, as shown by lower median finishing times and narrower interquartile ranges (IQRs). In contrast, female participants exhibit a wider spread in performance, with more extreme outliers, especially in the United States, Germany, and the United Kingdom. An exception to this trend is seen in Sweden, where female participants outperformed their male counterparts, standing out from the overall pattern.

# Save the updated dataset

write_excel_csv(df, "hyrox_results_updated_V1.csv") # Added new fields
write_excel_csv(hyrox_df, "hyrox_results_updated_V2.csv") # Time in seconds
write_excel_csv(filtered_df, "hyrox_results_updated_V3.csv") # Top countries for open
write_excel_csv(s5_filtered_df, "hyrox_results_updated_V4.csv") # Top countries for open S5